Proc. NatI. Acad. Sci. USAVol. 85, pp. 8810-8814, December 1988Biochemistry

Identification of a unique isoform of 1-aminocyclopropane-1-carboxylic acid synthase by monoclonal antibody

(ethylene/1-aminocyclopropane-l-carboxylic acid synthase/hormones/Lycopersicon esculentum L.)

ARKESH M. MEHTA, RAMON L. JORDAN, JAMES D. ANDERSON, AND AUTAR K. MATTOO*Beltsville Agricultural Research Center, U.S. Department of Agriculture, Agricultural Research Service, Beltsville, MD 20705

Communicated by Kenneth V. Thimann, August 5, 1988 (received for review May 31, 1988)

ABSTRACT 1-Aminocyclopropane-1-carboxylic acid(ACC) synthase (EC 4.4.1.14) is a key enzyme regulatingethylene biosynthesis in higher plants. A monoclonal antibody(mAb T20C) that immunoprecipitates the ACC synthase ac-tivity from tomato pericarp tissue extracts revealed that mAbT20C immunodecorates an =67-kDa polypeptide. On isoelec-tric focusing gels, ACC synthase activity in cell-free prepara-tions was resolved into three distinct activity peaks with pIvalues 5.3, 7, and 9. mAb T20C specifically recognized the pI7 form of the enzyme on electrophoretic transfer (Western)blots. When analyzed by sodium dodecyl sulfate gel electro-phoresis under reducing conditions, the eluted pI 7 form wasconfrmed to migrate as a polypeptide of 67 kDa. The 67-kDapI 7 isoform is a previously undescribed form ofACC synthase.

Ethylene is considered a plant hormone influencing manyaspects of growth, development, and senescence of higherplants (1, 2). The biosynthesis of ethylene from methionineoccurs by way of the following metabolic sequence: methio-nine -* S-adenosyl-L-methionine (AdoMet) -* 1-aminocyclo-propane-1-carboxylic acid (ACC) -- ethylene (for review seeref. 2). In addition to being developmentally regulated,ethylene biosynthesis is also promoted by a variety of envi-ronmental, chemical, and physicomechanical stimuli (1, 3). Acommon endogenous site where the effect of these stimuli ismore often realized is the step of conversion of AdoMet toACC catalyzed by the enzyme ACC synthase (EC 4.4.1.14)(2, 4). The mechanisms that underly induction or repressionof this enzyme during ripening and senescence of planttissues remain to be elucidated.The regulation of ACC synthase is of considerable interest

in plant hormone research. However, its low abundance andlabile nature have impeded molecular studies that can bedone with a purified protein. Furthermore, reports of itssubunit molecular mass range from 50 kDa in tomato fruit (5)to 72 kDa in potato (6) and 84 kDa in squash (7). These variedreports on the size of the protein have not been reconciled.By using protease inhibitors in the extraction and purificationbuffers and a different purification procedure, we havepartially purified a form ofACC synthase not described in theliterature and raised monoclonal antibody (mAb) to it. Thisunique ACC synthase protein is shown to have an apparentmolecular mass of 67 kDa on NaDodSO4/PAGE and a pI of7 on isoelectric focusing (IEF) gels. The denatured andundenatured enzyme form are recognized by the mAb.

MATERIALS AND METHODSPlant Material. Fruits from greenhouse-grown or field-

grown tomato plants (Lycopersicon esculentum cv. Pik-Red)were harvested at the early red stage, surface-sterilized with

70% ethanol, and thoroughly rinsed with sterile water. Thefruits were sliced, seeds were removed, and the pericarptissue was incubated for 8 hr to induce ACC synthase activityprior to homogenization as described (8) except that, inaddition, the homogenization buffer contained 0.8 mM phe-nylmethylsulfonyl fluoride and 100 units of aprotinin per ml.ACC Synthase Assay. Aliquots of the enzyme preparation

(5-200 ,ug of protein) were incubated for 1 hr at 30°C in 50mMN-[2-hydroxyethyl]piperazine-N'-3-propanesulfonic acid(EPPS)/KOH, pH 8.5, containing 50 ,uM AdoMet and 5 ,Mpyridoxal phosphate in a final volume of 1 ml. The enzymereaction was stopped with the addition of 100 ,l of 10 mMmercuric chloride. ACC formed was chemically converted toethylene, which was then quantified by gas chromatography(9). One unit of enzyme activity is defined as the productionof 1 nmol of ACC per hr at 30°C.

Purification of ACC Synthase. Routinely, crude homoge-nates prepared as described above were frozen in liquidnitrogen, Iyophilized, and stored at -80°C. The lyophilizedpowder was suspended in buffer A (1:5, wt/vol), whichcontained 2 mM EPPS/KOH, pH 8.5, 0.4 mM dithiothreitol,and S AM pyridoxal phosphate in addition to the followingprotease inhibitors: 0.5 mM phenylmethylsulfonyl fluoride,10 ,uM leupeptin, and 1 ,g of pepstatin A per ml. Thesuspension was clarified at 11,000 x g for 5 min. To thesupernatant, 0.5 vol of cold ethanol (-20°C) was added andthe mixture was stirred gently at -5°C for 15 min. Theprecipitates were centrifuged at 10,000 x g for 15 min anddiscarded. More ethanol (at -20°C) was added to the super-natant to obtain 50% saturation. The resulting precipitate,containing most of the enzyme activity, was collected bycentrifugation and dialyzed overnight against 2 liters of bufferA with four changes. The dialysate was applied to a SephadexG-100 column (bed volume, 276 ml) previously equilibratedwith buffer A. The fractions containing enzyme activity(specific activity, 12.1 units/mg of protein) were pooled andapplied to an ethyl agarose (Miles) column (bed volume, 30ml) equilibrated as above. Under these conditions the en-zyme binds to the column (8). The column was washed with10 vol of buffer A before applying a linear gradient of 10-300mM NaCI in buffer A to elute bound proteins. The enzymeeluted between 80 and 170 mM NaCl. The active enzymefractions (specific activity, 117 units/mg of protein) werepooled and gel filtered on a Sephadex G-100 column (bedvolume, 276 ml). The gel-filtered active fractions (specificactivity, 805 units/mg of protein) were concentrated byIyophilization and stored at -20°C. All enzyme purificationsteps were carried out in the cold room (4-8°C).

Abbreviations: ACC, 1-aminocyclopropane-1-carboxylic acid; IEF,isoelectric focusing; mAb, monoclonal antibody; AdoMet, S-adenosyl-L-methionine.*To whom reprint requests should be addressed at Plant MolecularBiology Laboratory, U.S. Department of Agriculture/AgriculturalResearch Service/Beltsville Agricultural Research Center, Bldg.006, Beltsville, MD 20705.

8810

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 25

, 202

0

Proc. Natl. Acad. Sci. USA 85 (1988) 8811

Protein Analysis and NaDodSO4/PAGE. Protein contentwas measured by the Bradford method (10) using Bio-Radreagent according to the supplier's manual. Proteins were

solubilized in sample application buffer (11), fractionated byNaDodSO4/PAGE according to Laemmli (12) using 5%stacking gel and 10-20% gradient or 12% polyacrylamide gelfor separation, and then stained with either Coomassie blueor silver (13, 14). Molecular masses (kDa) were determinedby comparison to 14C-labeled molecular mass (Amersham)and high molecular mass range protein (Bethesda ResearchLaboratories) standards.Immunization and Preparation of Hybridomas. A 12-

week-old pristane-primed (primed 10 days before the firstinjection with 0.5 ml of 2,6,10,14-tetramethylpentadecane)female BALB/c mouse was injected i.p. with 30 ,ag ofpurified ACC synthase (15 ug of native and 15 ,ug ofNaDodSO4/denatured protein) emulsified with Freund'scomplete adjuvant. Each subsequent injection was givenafter emulsifying the antigen in Freund's incomplete adju-vant. The mouse was reinjected with 1:1 native/denaturedantigen (total, 30 ,ug) 10 days after the first injection, with 50,.g of native antigen on day 21, and with 3:1 native/denaturedantigen (40,g; without the adjuvant) on day 32. The spleenwas removed on day 36. Mouse myeloma cell line P3/NS1/1-Ag4-1 was maintained in RPMI 1640 medium containing 13%fetal bovine serum, 1 mM sodium pyruvate, 2 mM L-glutamine, and 25 ,ug of gentamycin per ml (RPMI medium).Dissociated spleen cells from the immunized mouse werefused with NS1 myeloma cells at a 5:1 ratio using 45%(wt/vol) polyethylene glycol 4000 (Sigma) essentially as

described by Oi and Herzenberg (15). Hybrid cell lines wereplated at low cell density and initially maintained in RPMImedium containing 20% myeloma-conditioned medium(spent culture supernatant from 48-hr myeloma cultures;filtered), 6% Nu-serum (Collaborative Research, Waltham,MA), 6 mM Hepes, 0.018 mM 2-mercaptoethanol, 0.1 mMhypoxanthine, 0.016 mM thymidine, and 0.004 ,uM aminop-terine (fusion plating medium). Once established, hybrid-omas were grown in fusion plating medium without amino-pterin.

Production and Selection of mAbs. Hybridomas secretingantibodies against ACC synthase were selected by testing theculture supernatants using an indirect ELISA, immunodotblot, and immunoprecipitation assay. Antigen-specific, anti-body-secreting hybridomas were cloned by limiting dilution.Antibody-containing ascites fluid was produced by injecting107 hybridoma cells i.p. into 12- to 20-week-old BALB/c miceprimed 10-20 days previously with 0.5 ml of pristane. mAbswere concentrated from ascites fluid by sequential (25% and50%) ammonium sulfate fractionation (retaining the 25-50%fraction; ref. 16). Antibodies were further purified by proteinA-Sepharose chromatography essentially as described by Eyet al. (17). Antibody class and subclass were determined byOuchterlony double-diffusion using isotype-specific rabbitantisera (Litton Bionetics).

Antibodies not reactive to ACC synthase were used asnegative controls. These included a mAb (an IgG1; desig-

nated SC 4G4GS) reactive to a cell membrane protein on theplant pathogenic corn stunt spiroplasma (Spiroplasmakunkeli) (18) and a mAb (an IgM; designated BYMV-ICN7C3) reactive to the large subunit of ribulose bisphosphatecarboxylase (R.L.J., unpublished).Dot blot analysis (19) was used as the initial screen for

selecting cell lines secreting ACC synthase-specific antibod-ies.

Immunoprecipitation Assay and Electrophoretic Transfer(Western) Blotting. The ability of antisera or antibodies toinhibit tomato ACC synthase activity was determined byincubating 5-25 units of partially purified enzyme in 250-A1aliquots with increasing amounts of antibody diluted inTris-buffered saline (pH 8.0) for 4-16 hr at 40C. The precip-itate obtained was centrifuged at 10,000 x g for 10 min at 4°C,and the supernatant was assayed for residual enzyme activ-ity. Control tubes contained preimmune serum or purifiedantibodies to unrelated antigens. Fifty microliters of proteinA-Sepharose CL-4B (Pharmacia) or 25 Al of goat anti-mouseIgG immunobeads (Bio-Rad, 10 mg/ml) was added to eachtube and incubated at 25°C for 2 hr. The precipitates werecentrifuged and the supernatant was assayed for any residualenzyme activity. For Western blot analysis, proteins weretransferred from 10-20% gradient or 12% polyacrylamidegels onto nitrocellulose paper (Schleicher & Schuell) andprobed with antisera according to Roberts et al. (20) usinggoat anti-mouse alkaline phosphatase conjugate or 125[_labeled protein A (DuPont/NEN) as the secondary antibody.IEF and Western Blotting. IEF of the homogenates was

performed according to the Pharmacia technical manualusing 6.5% Ampholine/agarose gels (pH 3.5-10), with a

focusing time of 1800-2000 V hr. After completion of the run,the pH gradient was determined by comparing migration ofknown isogel pI standard markers as described in the FMCBioproducts technical bulletin. The gels were fixed for 10 minin 20% trichloroacetic acid, stained for 15 min with Serva blueW, and destained in water. Identical gels were transferred tonitrocellulose papers by passive diffusion for 2 hr under a

pressure of 1 kg weight. Following the transfer, the nitrocel-lulose papers were air dried for 1 hr and then treated withantiserum and detected with 125I-labeled protein A. Foreluting proteins from agarose gels, the gels were sliced into40 strips, each 5 mm. The strips were incubated overnightwith buffer A at 4°C with gentle shaking. The eluted proteinswere gel-filtered using Sephadex G-25 spun column chroma-tography (21) before determining ACC synthase activity,protein content, and NaDodSO4/PAGE pattern.

RESULTS

Purification of ACC Synthase. ACC synthase was purified900-fold from tomato pericarp extracts following a protocol(described in Materials and Methods) that includes a numberof protease inhibitors in all buffers, ethanol precipitation,hydrophobic chromatography, and gel filtration. Data fromthe various purification steps are summarized in Table 1. Thepurified protein had a specific activity of 805 nmol/hr per mg

Table 1. Summary of purification of ACC synthase from tomato fruit

Specific activity,Total % nmol/hr per mg Fold

Purification step units recovery of protein purification

Homogenate 262 100 0.91 1Ethanol precipitate 125 48 0.65 0.8Sephadex G-100 gel filtration (I) 44 17 12.1 13Ethyl agarose eluate 63 24 117 130Sephadex G-100 gel filtration (II) 22 8 805* 900

*This partially purified enzyme preparation had Km values of 14.5 and 0.35 AM for AdoMet andpyridoxal phosphate, respectively. It was inhibited by aminoethoxyvinylglycine with a Ki of 2.2 AuM.

Biochemistry: Mehta et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 25

, 202

0

Proc. Natl. Acad. Sci. USA 85 (1988)

a1)0Coo xo)1 a)< V

co

c:la)LL U)

FIG. 1. NaDodSO4/PAGEpattern of proteins at differentstages of purification of ACCsynthase (Table 1). Twenty-fivemicrograms of protein of the ho-mogenate and the ethanol pre-cipitate (ppt.) and 5 ,g of proteinof the remaining fractions weresubjected to electrophoresis un-der denaturing conditions on 10-20%o gradient NaDodSO4/poly-acrylamide gels. After fixation,the gels were stained with silver.The positions of the molecularmass protein standards are indi-cated in kDa.

of protein. However, we consistently observed that theenzyme is slowly inactivated upon isolation and duringpurification; therefore, the observed specific activity andpurification level are underestimates. The NaDodSO4/PAGEprofile of proteins during each stage ofenzyme purification isshown in Fig. 1. The 900-fold purified fraction was enrichedin three or four major components with molecular masses of43, 53, 67, and 94 kDa (Fig. 1, Sephadex II; Fig. 3A, lane 1).One or more of these polypeptides could be either bona fidesubunits of ACC synthase, degradation products of theenzyme, or contaminants. Gel filtration on Sephadex G-100or on Superose columns indicated a native molecular mass of-65 kDa for the active enzyme. Therefore, the 67-kDa

100

._

0cc 80Co0

C.

u) 6000

2 400

20

0 50 100 150 200 250

Antibody (ng/ACC synthase unit)

FIG. 2. Immunotitration of ACC synthase activity in SephadexI fraction (Table 1). The residual enzyme activity was determinedafter addition of various amounts of mAb T20C (o) or mAb SC4G4GS prepared against an unrelated protein (o). The enzymeactivity of the untreated control was 1.9 units/ml. The titer of theantibody was found to be 25 ,ug/unit of ACC synthase activity.

protein detected in NaDodSO4/polyacrylamide gels wasconsidered a prime candidate for a monomeric ACC syn-thase.Immunochemical Characterization ofACC Synthase. mAbs

were raised against the Sephadex II fraction. Fifty positivehybridoma culture supernatants were assayed for secretionof antibodies that inhibited and immunoprecipitated theactivity of ACC synthase from partially purified prepara-tions. Seven stable hybridoma cell lines secreting antibodieswere selected. One hybridoma was subcloned and antibodiesfrom its ascites fluid were purified. This mAb was found tobe IgG1 by immunodiffusion assay and was designated mAbT20C. Immunoprecipitation of ACC synthase activity bymAb T20C followed saturation kinetics (Fig. 2). Normalmouse serum, ascites fluid from the myeloma cell line, andantibodies purified from ascites fluid containing mAbs to anunrelated plant pathogen membrane protein (Fig. 2, controlmAb SC 4G4GS) did not inhibit or immunoprecipitate theenzyme activity. The material immunoprecipitated by mAbT20C was devoid of any ACC synthase activity. Also, it wasnoted that mAb T20C immunoprecipitated between 78% and92% of the enzyme activity from partially purified prepara-tions and only 47% and 63% of that in cell-free extracts.The specificity of the mAb was demonstrated by Western

blot analysis. Fig. 3A, lane 2, shows that mAb T20C immu-nodecorates a polypeptide that coelectrophoreses with the67-kDa polypeptide in the enzyme preparation used as an

A B

97.4 -

68 -

43 -

25.7 -

18.4 -

1 2 1 2

FIG. 3. (A) Immunodetection of67-kDa protein by mAb T20C onWestern blot (lane 2) of Sephadex II fraction (lane 1, silver stained).Five microgram protein equivalents of Sephadex II fraction wasresolved by NaDodSO4/PAGE. One part of the gel was transferredonto nitrocellulose paper. The nitrocellulose blot was incubated withmAb T20C and the antigen-antibody complex was detected by usinggoat anti-mouse alkaline phosphatase conjugate (lane 2). The otherhalf of the gel was stained with silver (lane 1). Molecular masses aregiven in kDa. (B) Detection of low molecular mass protein products(indicated in kDa) on immunoblots. Cell-free extracts were preparedin a buffer containing 50 mM Tris HCI (pH 8.5), 20 ,uM magnesiumsulfate, 1 mM EDTA, 5 ,uM pyridoxal phosphate, 5 mM 2-mercaptoethanol, 10 ,uM leupeptin, 1 ,tg of pepstatin A per ml, 0.5mM phenylmethylsulfonyl fluoride, and 10% glycerol with (lane 2)and without (lane 1) 50 mM NaCl. Twenty-five micrograms of proteinof each extract was fractionated by NaDodSO4/PAGE, transferredonto nitrocellulose paper, and probed with mAb T20C.

0

ao a

o c m

E ,: Qo co 0

I LUJ U)

97.4 -SW68 - 3

43 - 1

25.7 - l

18.4 - 4 *m

14.3 - i i.... ..

8812 Biochemistry: Mehta et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 25

, 202

0

Proc. Natl. Acad. Sci. USA 85 (1988) 8813

immunogen. A 67-kDa polypeptide present in tomato cell-free extracts was similarly identified by mAb T20C (Fig. 3B,lane 1). However, when cell-free extracts were prepared inbuffers containing salt (Fig. 3B, lane 2) or repeatedly frozenand thawed prior to electrophoresis and Western blotting(data not shown), several low molecular mass polypeptides(47 kDa, 38.5 kDa, 37-kDa doublet, and 31 kDa) reacting withthe antibody were revealed (compare Fig. 3B, lane 2, withlane 1). This immunodecoration of low molecular mass poly-peptides apparently occurred at the expense of the 67-kDaprotein band, suggesting its labile nature. However, theidentity ofthe smaller polypeptides as breakdown products ofthe parent 67-kDa protein remains to be elucidated.Ioforms ofACC Synthase. Preliminary observations on the

presence of isoforms of ACC synthase (22) together withvaried reports (see Introduction) on the size of this enzymeprompted us to check if mAb T20C would recognize one ormore of these isoforms. Cell-free extracts were fractionatedon IEF agarose gels. The IEF gels were sliced and proteinseluted from them were assayed for ACC synthase activity. Inthis way, the cell-free extracts were resolved into threedistinct enzyme activity peaks focusing at pI values 5.3, 7,and 9 (Fig. 4). The predominance of the pI 7 isoform was alsoapparent in pH-dependent binding assays (23). Identical butunsliced IEF agarose gels were blotted onto nitrocellulosepaper and immunodecorated with mAb T20C. A minor

I I NO I 0 14141 v.N I_.

&O

10

8

!6

I

4

2

I] Stain

I

20

16 E

Ct

c

._

zu

12

co(D

(n

8 m

n4 )

L)

44<

0 10 20 30 0

Fraction number

FIG. 4. Multiple molecular forms of ACC synthase in tomatofruit extracts and selective reactivity of pI 7 form with mAb T20C.Two hundred micrograms of protein of the cell-free extracts wasfractionated under native conditions on IEF agarose gels with a pHgradient between 3.5 and 10. One part of the gel was stained withServa blue W (Stain) and the second part was transferred to nitro-cellulose paper and immunodecorated with mAb T20C (Blot). Theremainder of the gel was sliced into 5-mm strips, proteins were elutedby an overnight incubation at 4°C with buffer A, and the eluates wereassayed for ACC synthase activity. Bars represent the enzymeactivity. The separation of protein standards (M) on the IEF gel isalso shown. Three isoforms of enzyme activity apparent at pI values5.3, 7, and 9 comprise 30%o, 53%, and 17%, respectively, of totalactivity.

Stain

69 -

46 -

30 --

Immunoblot

4-

0I

14.3-

5.3 7 9 5.3 7 9

PI

FIG. 5. NaDodSO4/PAGE pattern and Western blot analysis ofprotein fractions with pI values 5.3, 7, and 9. Proteins eluted fromIEF gels at the indicated pI values were separated on 10-20%ogradient NaDodSO4/polyacrylamide gel in duplicate. One part wasstained with silver (Stain); the remaining part was transferred tonitrocellulose paper and immunodecorated with mAb T20C (Immu-noblot) using 1251-labeled protein A. Positions of 14C-labeled proteinmolecular mass standards (indicated in kDa) and the 67-kDa protein(arrow) are given.

protein (see stained part of the gel) that cofocused with the pI7 enzyme activity peak was revealed on the Western blot(Fig. 4). To further ascertain the relationship between the pI7 form of ACC synthase and the 67-kDa protein, proteins infractions that focused in IEF gels at pI values 5.3, 7, and 9were subjected in duplicate to second-dimension NaDod-S04/PAGE. One part was silver stained, whereas the otherhalf was blotted onto nitrocellulose paper and challengedwith mAb T20C. The results, shown in Fig. 5, indicate thatthe pI 7 isoform of ACC synthase resolves into a 67-kDaprotein that is recognized on Western blots by mAb T20C.

DISCUSSIONBy using a mAb we have identified a previously undescribedform of ACC synthase present in ripe tomato fruit. Theantibody immunoprecipitated the enzyme activity from cell-free extracts and recognized a single protein band of 67 kDaon Western blots. Gel filtration of the partially purified,native tomato enzyme revealed a molecular mass of about 65kDa. These data in conjunction with the NaDodSO4/poly-acrylamide gel analysis under denaturing conditions indicateACC synthase to be a monomeric enzyme of about 67 kDa.The purification procedure used in our study was designed tominimize 'proteolytic degradation that may occur duringenzyme isolation and avoid the use of ammonium sulfate forprecipitation/concentration of the enzyme. In our hands,greater losses and enzyme inactivation were commonlyobserved whenever ammonium sulfate was used to salt outthe enzyme. Recently Bleecker et al. (5) reported a mAb thatinhibited ACC synthase activity in cell-free extracts. Whenan affinity column with this antibody was used to bind totomato fruit proteins, a protein of 50 kDa was eluted fromsuch a column under denaturing conditions (5, 24). However,their mAb could not be used on immunoblots. By a differentradiolabeling approach, another study implicated a 50-kDaprotein as a putative ACC synthase (25). Both studies usedammonium sulfate precipitation as an initial step in theirpurification protocols.

Biochemistry: Mehta et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 25

, 202

0

Proc. Natl. Acad. Sci. USA 85 (1988)

It is possible that the differences in the molecular massesreported from several laboratories are due to the differentpurification procedures used. In this context, it is of partic-ular interest to note that another study which did not employan ammonium sulfate precipitation step in the purification ofACC synthase (from etiolated mung bean hypocotyl seg-ments) found the enzyme to be a dimer of identical subunitsof about 65 kDa (26). The initial report of a molecular massof 84 kDa for the enzyme from squash (7) seems to have beenin error. More recent immunological evidence suggests thesquash enzyme to be a dimer of identical subunits of 60 kDa(27). Furthermore, it should be noted that the presence of saltduring preparation of homogenates, or repeated freezing andthawing of cell-free extracts, can lead to detection ofpossiblydegraded forms of the parent protein and therefore toartifactual results (Fig. 3).

In a complementary experiment (28), we utilized therequirement of pyridoxal phosphate and its association withACC synthase as a cofactor (2) to radiolabel the enzyme withsodium [3H]borohydride by reduction of the Schiff baseformed between the enzyme and the cofactor (25, 29). Inthese experiments as well, a 67-kDa protein was labeled. The3H-labeled 67-kDa protein coelectrophoresed with the en-riched 67-kDa protein band in the purified ACC synthasefraction and was recognized by mAb T20C. Thus, thepyridoxal phosphate data support the contention that the67-kDa protein reported here is ACC synthase.On IEF agarose gels, tomato fruit extract was resolved into

three isoforms of ACC synthase having pI values 5.3, 7, and9. The pI 7 and pI 9 forms as major enzyme activity peakswere previously reported by us (22). The pI 5.3 form of ACCsynthase may be related to a pI 4.5 form that was recentlyreported (25). The mAb T20C recognized the pI 7 form onWestern blots of the IEF agarose gels. In addition, the pI 7form was resolved into a 67-kDa protein on NaDodSO4/poly-acrylamide gels that reacted with mAb T20C on Westernblots. Therefore, mAb T20C recognizes the native as well asthe denatured form of pI 7 ACC synthase but, within thedetection limits, not any other isoform or their denaturedmonomers. These data suggest that this mAb recognizes anepitope on ACC synthase protein which is probably notshared by other forms of the enzyme. It needs to bedetermined if mAb T20C can recognize related proteins fromdifferent plant sources.The demonstration of different forms of ACC synthase

raises several interesting questions concerning the develop-mental and environmental regulation of the enzyme. Are thedifferent isoforms products of different genes? Are theseisoforms interconvertible? Do different isoforms arise inresponse to different stimuli? For instance, is wound-inducible ACC synthase different from the developmentallyregulated enzyme? It may be pertinent to point out that of thegenes whose transcription is induced by ethylene-e.g.,cellulase (30), polygalacturonase (31), and chitinase (32)-allare present either as multiple gene families or have isoenzy-mic forms. It will be interesting to know if the genes thatencode the mRNAs for the biosynthetic enzymes for ethy-lene-for instance, ACC synthase-also belong to multiplefamilies. To start answering these questions, the mAb can beused to analyze ACC synthase expression at the gene leveland isolate the cognate gene. Such an approach should helpin defining the relationship between the different ACCsynthase isoforms and the molecular regulation of theirexpression.

We thank Drs. Marvin Edelman, Jeffrey Dean, Sudhir Sopory, andJeffrey Suttle for a critical reading of the manuscript, Mary AnnGuaragna and Marcia Sloger for excellent technical assistance, andRoshni Mehta for help with illustrations.

1. Lieberman, M. (1979) Annu. Rev. Plant Physiol. 30, 533-591.2. Yang, S. F. & Hoffman, N. E. (1984)Annu. Rev. Plant Physiol.

35, 155-189.3. Abeles, F. B. (1973) Ethylene in Plant Biology (Academic, New

York), pp. 30-102.4. Mattoo, A. K. & Aharoni, N. (1988) in Senescence and Aging

in Plants, eds. Nooden, L. D. & Leopold, C. A. (Academic,New York), pp. 241-280.

5. Bleecker, A. B., Kenyon, W. H., Somerville, S. C. & Kende,H. (1986) Proc. Natl. Acad. Sci. USA 83, 7755-7759.

6. Bums, J. K. & Evensen, K. B. (1986) Physiol. Plant. 66, 609-615.

7. Nakajima, N. & Imaseki, H. (1986) Plant Cell Physiol. 27, 969-980.

8. Mattoo, A. K., Adams, D. O., Patterson, G. W. & Lieberman,M. (1983) Plant Sci. Lett. 28, 173-179.

9. Lizada, M. C. C. & Yang, S. F. (1979) Anal. Biochem. 100,140-145.

10. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254.11. Mattoo, A. K., Pick, U., Hoffman-Falk, H. & Edelman, M.

(1981) Proc. Natl. Acad. Sci. USA 78, 1572-1576.12. Laemmli, U. K. (1970) Nature (London) 227, 680-685.13. Fairbanks, G., Steck, T. L. & Wallach, D. F. H. (1971) Bio-

chemistry 10, 2606-2617.14. Merril, C. R., Goldman, S., Sedman, S. A. & Ebert, M. H.

(1981) Science 211, 1437-1438.15. Oi, V. T. & Herzenberg, L. A. (1980) in Selected Methods of

Cellular Immunology, eds. Mishell, B. B. & Shiggi, S. M.(Freeman, San Francisco), pp. 131.

16. Parham, P. (1983) Methods Enzymol. 92, 110-138.17. Ey, P. L., Prause, S. J. & Jenkin, C. R. (1978) Immunochem-

istry 15, 429-436.18. Jordan, R. L., Konai, M., Lee, I. M., Hsu, H. T. & Davis,

R. L. (1987) in Plant Pathogenic Bacteria, eds. Civerolo,E. L., Collmer, A., Davis, R. E. & Gillaspie, A. G. (Nijhoff,Boston), pp. 340-346.

19. Jahn, R., Schiebler, W. & Greengard, P. (1984) Proc. Natl.Acad. Sci. USA 81, 1684-1687.

20. Roberts, D. R., Thompson, J. E., Dumbroff, E. B., Gepstein,S. & Mattoo, A. K. (1987) Plant Mol. Biol. 9, 343-353.

21. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) MolecularCloning: A Laboratory Manual (Cold Spring Harbor Labora-tory, Cold Spring Harbor, NY).

22. Mattoo, A. K. & Anderson, J. D. (1984) in Ethylene: Biochem-ical, Physiological and Applied Aspects, eds. Fuchs, Y. &Chalutz, E. (Nijhoff, The Hague), pp. 139-147.

23. Yang, V. C. & Langer, R. (1987) BioTechniques 5, 138-144.24. Bleecker, A. B., Robinson, G. & Kende, H. (1988) Planta 173,

385-390.25. Privalle, L. S. & Graham, J. S. (1987) Arch. Biochem. Biophys.

253, 333-340.26. Tsai, D., Arteca, R. N. & Phillips, A. T. (1987) Plant Physiol.

Suppl. 83, 90 (abstr.).27. Imaseki, H., Nakajima, N. & Todaka, I. (1988) in Biomecha-

nisms Regulating Growth and Development, eds. Steffens,G. L. & Rumsey, T. S. (Kluwer, Dordrecht, The Netherlands),pp. 205-227.

28. Mehta, A. M., Anderson, J. D. & Mattoo, A. K. (1988) FASEBJ. 2, A1024 (abstr.).

29. Lundblad, R. L. & Noyes, C. M. (1984) Chemical ReagentsforProtein Modification (CRC, Boca Raton, FL), pp. 142-147.

30. Christophersen, R. (1987) in Plant Senescence: Its Biochemis-try and Physiology, eds. Thomson, W. W., Nothnagel, E. A. &Huffaker, R. C. (Am. Soc. Plant Physiol., Rockville, MD), pp.89-97.

31. Ali, Z. M. & Brady, C. J. (1982) Aust. J. Plant Physiol. 9, 155-169.

32. Broglie, K., Durand-Tardiff, M., Gaynor, J. J. & Broglie, R.(1986) Proc. Natl. Acad. Sci. USA 83, 6820-6824.

8814 Biochemistry: Mehta et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 25

, 202

0